Download offer better outcomes

U.O. di Nefrologia, Dialisi ed Ipertensione
Policlinico S.Orsola-Malpighi
Azienda Ospedaliero-Universitaria
Bologna - ITALY
Do advances in hemodialysis technology
(e.g.the use of biofeedback, blood volume and
clearance monitoring) offer better outcomes ?
Pro : Antonio Santoro M.D.
Dialysis related complications in Conventional HD
Hypertension
•Mittal SK, Clin Nephrol,1999, 51 (2), 77-82
•Chazot C et al, ND &T, 1995, 10, 831-837
Hypotension
•De Santo G et al, AJKD, 2001, 38 (4) S1, S38-S46.
•Agarwal et al, AJKD, 2008, 51, (2), 242-254
Vascular access
failure
•Reyner HC et al, Kidney Int, 2003, 63 , 323–330
(Probability of vascular access
failure within one year)
Cardiac arrhythmias
1.1
Narula AS, Ren Fail. 2000
Abe S Am Heart J. 1996
Micro.macro
inflammation
•Stenvinkel P, Kidney Int., Vol. 62 (2002), 1791–1798
Hospitalization
•Rayner HC et al, ND & T (2004) 19: 108–120.
•O’Brien T, AJKD, 2008, 51 (1), S1, S137-154
(Rate per patient-year)
Intra-dialytic Hypotension may….
Interfere with the delivery of adequate
dialysis
Induce or aggravate hypoperfusion in
different districts:
cerebral
mesenteric
cardiovascular
Influence the patient outcome
Hemodynamic
instability and outcome
The effect of frequent or occasional
dialysis-associated hypotension on
survival of patients on maintenance
hemodialysis
Tisler A, NDT 2003
Hemodynamic stability
during HD
has to be considered a dialysis
ADEQUACY PARAMETER
Shoji T, Kidney Int 2004
Hemodialysis-induced myocardial stunning
PET, for myocardial
perfusion assessment
Mc Intyre, CJASN 2008
Hemodialysis-induced myocardial stunning
HEMODIALYSIS
↓ Myocardial blood flow
Myocardial ischemia
Regional wall motion abnormalities
Myocardial dysfunction
Towards a more physiological dialysis
Daily frequency
Long duration
Incremental lifetime cost ($)
Incremental cost and incremental life expectancy
relative to current practice under baseline
assumptions.
The cost-effectiveness ratio increases with the frequency of hemodialysis.
More frequent in-center hemodialysis strategies could become cost-neutral if
the cost per hemodialysis session could be reduced by 32 to 43%.
Incremental Life Expentancy
(months)
Lee CP. et al. JASN 2008;19:1972
Towards a more physiological dialysis
Daily frequency
Long duration
Alternatively or combined with :
high technology
Control of dialysate chemical and physical
properties (temperature)
Control of dialysis efficiency
Monitoring and control ( in open or closed loop)
of the hemodynamic patient variables: SAP,
TPVR, BV, HR, CO
Tailoring of ultrafiltration and conductivity by
means of feedback control systems
DIALYSIS-RELATED
CARDIOVASCULAR INSTABILITY
NON AUTONOMIC
CAUSES
BV volume
depletion
Myocardial
insufficiency
Stroke vol
• Eccessive UFR
• TBW/dry weight < 50%
• Negative Na balance
• Maldistribution of BV
AUTONOMIC
CAUSES
Vaso
regulatory
impairment
Venodilation
Decrease in PVR
Vaso
vagal reflex
• Bezold Jarish
Secondary
Autonomic
failure
• Baroreceptor
deafferantation
• Sympathetic
disfunction
CONTINUOUS METHODS FOR MEASURING
RELATIVE BLOOD VOLUME
Optical absorption of
monochromatic light
Sound speed
Conductivity
Viscosity
~
Intensity
Haemoglobin
Haematocrit
Hemoconcentration
Sound speed
Conductivity
Differential pressure
Total
protein
Hemoconcentration
Hemoconcentration
Hemoconcentration
Mass conservation principle ; blood substances confined to the vascular space
change proportionally as a result of changes of the plasma volume
d BV = - [F + UF] + R
A(t)
V(t)
dt
0
Relative BV changes
Increasing
vasoconstriction
-10
Decreasing
tissue
hydration
-20
-30
0
60
120
180
240
TIME (mins)
Plasma Volume
Change rate
=
Vascular refilling
rate
-
Ultrafiltration
rate
Relationship BV - UFR
Blood volume change, %
Ultrafiltration rate L/h
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
0
-5
-10
-15
-20
-25
-30
-35
y=-9.323x-6.824
r=-0.33;p=0.007
n=102
Mancini et al. IJAO, 1996
Hematocrit threshold determined by the Crit-Line Instrument
When Intradialytic Morbid Events Occurred
Patient
No.
No. Of
IME*
1
2
3
4
5
6
7
8
9
10
11
12
9
6
7
4
2
3
9
3
6
3
2
7
Hematocrit Threshold
(mean ±SD)
40
38
31
46
44
38
44
40
38
49
36
49
±
±
±
±
±
±
±
±
±
±
±
±
1.6
1.3
2.7
1.9
0.5
2.2
0.5
1.2
0.4
0.9
0.5
2.2
Mean H change before
IME from start
of treatment
8
4
4
12
10
8
6
4
5
11
6
9
* Total number of IME occurring during the six study sessions for each patient
Steuer R. et al, ASAIO J, 1994
An individual RBV limit exists for nearly all patients and this treshold may
mark the individual window of haemodynamic instabilities
MECHANISMS WHICH CAN AFFECT PLASMA
REFILLING DURING DIALYSIS
1.
Impairment of peripheral vasoconstriction during volume removal
•
•
•
•
2.
Acetate
Release of cytokines (IL1-TNF-IL6)
Autonomic neuropathy
Thermal stress
Increase in hydrostatic capillary pressure
• Compromised cardiac function
• Peripheral pooling of blood volume
3.
Depletion of interstitial volume
• Low dialysate sodium concentration
• High transcellular urea gradient
• Dry body weight error or high UF
4.
Oncotic pressure changes
• Hypoalbuminemia
• Alteration in interstitial fluid drainage and lymph flow
Blood volume changes
in normo- over- and under-hydrated patients
p<0.001
PRE - TO - POST HD
BLOOD VOLUME CHANGES
%
p<0.001
0
In under-hydrated pts. the ratio between
-8
BV% changes
and ultrafiltration is very
high, on the contrary in over-hydrated
pts. the -16
same ratio is very low
-24
Normohydrated
patients
Overhydrated
patients
Underhydrated
patients
The monitoring of RBV trends alone
may be misleading and confusing
J Am Soc Nephrol 2005; 16: 2162–2169
443 HD patients randomized to 6 months to Crit-line
conventional monitoring (227) or conventional
monitoring (216)
Results
More non-access-related hospitalizations were seen in the BVM
compared with conventional groups (120 vs 81 episodes)
The unadjusted and adjusted risk ratios for non-access-related
hospitalization were 1.49 and 1.61 respectively
The adjusted risk ratios for cardiovascular admissions was
1.85
The mortality at 6 months was greater in the BVM than the
conventional monitoring (8.7% vs 3.3%)
Limitations to the study
The study population was not limited to those with clinical issues of
volume management and hemodynamic instability
Changes
in the
without
any
Study patients
had a profiles
mean age of 59.2
yr
correlations
with UF and body weight of the
Comorbid conditions such peripheral vascular disease (19%) , left
ventricular
dysfunction
(11%), left ventricular
hypertrophy (25%) were
patients
were
intended
to
support
very low in frequency in comparison with the prevalent HD pts. in USA
and Europe
modifications
in the target post-dialysis weight
and/or
antihypertensive
medications
The conventional
monitoring group had
a lower than expected mortality
and hospitalization rate , which may exacerbate the differences between
the two groups
Cause of hospitalization was not centrally adjudicated
The study period only 6 months, a longer horizont might have different
findings
Summary of randomized BVM and BVT trials,
characterized by study population
Study
Study
design
Intervention
Primary
end point
Outcome
Conventional BV monitoring
Non-IDH-prone
Population
RCT
Crit-line (conventional
monitoring)
Morbidity
Increased hospitalization in
BVM group, Adjusted RR 1.61
(95% CI 1.15–2.25)
Fresenius 4008HD
(conventional monitoring)
Intra-dialytic
morbidity
Reduction in IDH in BVM group
(24% vs 32%, P = 0.04)
Reddan (JASN 2005)
Gabrielli (JN 2009)
Crossover
RCT
Blood volume tracking (closed loop) haemodialysis ( HemoBiofeedback)
IDH-prone population
Santoro (KI 2002)
Crossover
RCT
BVT (conventional HD)
IDH reduction
30% reduction in IDH, (P =0.004)
sessions in BVT Group
Ronco (KI 2000)
Crossover
RCT
BVT (conventional HD)
IDH reduction
Less IDH in BVT group (24 vs 59
HDx sessions, P< 0.001)
Hemo-biofeedback systems
(conventional monitoring)
Change in ECV
at 6 months
Lower IDH in HBS group (0.13
vs 0.31) P = 0.04)
Nersallah ( ASAIO J
2008)
RCT
READING BLOOD VOLUME & BLOOD PRESSURE
DURING HD
BV -7.5%
UF (L)
2500
3500
160
3000
140
2000
120
1500
100
1000
BV - 8.6%
80
BV - 9.1%
500
BV - 4.9%
BV - 5.1%
BV - 5.0%
180
160
2500
UF (L)
3000
180
Syst art press (mmHg)
BV - 6.2%
3500
RBV behaviour in overhydrated pt.
140
2000
BV - 4.8%
120
1500
100
1000
80
500
0
60
0
1
2
Treatment time (h)
3
4
0
60
0
1
2
Treatment time (h)
3
4
Syst art press (mmHg)
RBV behaviour in normohydrated pt.
Clinical significance of Monitoring the
Blood Volume variations
2
1
1.8
3
1.6
5
1.4
7
1.2
9
1
0.8
11
0.6
13
0.4
UF stop
17 0
40
80
120
Time (min)
160
240
200
160
120
80
40
-
15
200
Weight loss rate (Kg/h)
Blood volume change (%)
Assesment of plasma refilling rate
0.2
240
0.0
“The use of dynamic test, based on ultrafiltration stops, may be useful for optimising
the patient’s dry-weight and to evaluate the individual capillary filtration coefficient.”
Santoro A et al, Int. J Art. Org, 1997.
Chi-Square Analysis Comparing UF and Refill
Characteristics Between Hypotensive and
Normotensive UF Pulses
Parameter
Hypotensive
UF Pulses (n=30)
Normotensive
UF Pulses (n=60)
P
RBV at UF pulse initiation (%)
90.5+4.2
94.6+3.5
<0.001
∆RBVUF(%)
7.4+1.8
6.9+1.5
NS
UFVS(mL)
457+123
457+123
NS
0.017+0.005
0.016+0.004
NS
UF decay amplitude (b)
71.2+12.9
79.4+13.9
0.007
τUF
21.6+8.5
12.8+2.8
<0.001
Linear divergence (%.s)
155+285
662+405
<0.001
IRR (%/min)
0.86+0.45
0.76+0.35
NS
3.4+1.1
4.8+2.4
0.001
0.25+0.69
0.14+0.10
NS
∆RBVUF/UFVS(%/mL)
Refill phase amplitude
τref
Mitra S; Am J Kidney Dis 2002;40:556-565
Covariates related to symptomatic hypotension
multivariate logistic regression
Group
p
Relative
Risk
95% CI
<0.001
C vs A
NS
1.25
0.54-2.89
B vs A
<0.001
7.26
3.07-17.13
Baseline plasma-dialysate Na+ gradient
(for each 1 mEq/L increase)
<0.001
1.13
1.06-1.22
∆ BV from 20 to 40 min of dialysis (for
each 1% decrease)
0.030
1.23
1.02-1.48
Irregularity of BV over time (yes/no)
0.001
3.13
1.65-5.96
HR decrease from the start to the 20th
min of dialysis (for each 1 beat/min
decrese)
0.017
0.95
0.91-0.99
Andrulli S, Am J Kidney Dis 2002
Blood volume tracking
SYSTEM
∆%RBV
Patient
DC
WLR
∆%BV Target
BV
Monitor
DC
Monitor
BWL.
Monitor
DCeq target
BWL
Target
Errors
Dialysis
Machine
MIMO
Controller
Santoro A et al. Blood Volume Regulation During Hemodialysis,
Am J Kidney Dis, 1998
Blood Volume Tracking
0
TWL (L)
-5
BV (%)
5
-10
3
2
1
0
0
60
120
180
240
17
15
WLR (L/h)
13
DC (mS/cm)
time (min)
1,25
1
0,75
0,5
0,25
0
0
60
120
180
240
time (min)
Santoro A. AJKD1998
The first experiences with the biofeedback
control of Blood Volume
Time (min)
30
60
90
120
150
180
210
0
SAP changes (%)
-4
Biofeedback HD
-8
-12
-16
Standard HD
-20
*
-24
-28
*
*
* p<0.05 B vs A
Effects of automatic blood volume control over intradialytic hemodynamic stability, E. Mancini et
al, Int, J. Art. Org., 1995, 18, 9: 495-498
Summary of randomized BVM and BVT trials,
characterized by study population
Study
Study
design
Intervention
Primary
end point
Outcome
Conventional BV monitoring
Non-IDH-prone
Population
RCT
Crit-line (conventional
monitoring)
Morbidity
Increased hospitalization in
BVM group, Adjusted RR 1.61
(95% CI 1.15–2.25)
Fresenius 4008HD
(conventional monitoring)
Intra-dialytic
morbidity
Reduction in IDH in BVM group
(24% vs 32%, P = 0.04)
Reddan (JASN 2005)
Gabrielli (JN 2009)
Crossover
RCT
Blood volume tracking (closed loop) haemodialysis ( HemoBiofeedback)
IDH-prone population
Santoro (KI 2002)
Crossover
RCT
BVT (conventional HD)
IDH reduction
30% reduction in IDH, (P =0.004)
sessions in BVT Group
Ronco (KI 2000)
Crossover
RCT
BVT (conventional HD)
IDH reduction
Less IDH in BVT group (24 vs 59
HDx sessions, P< 0.001)
Hemo-biofeedback systems
(conventional monitoring)
Change in ECV
at 6 months
Lower IDH in HBS group (0.13
vs 0.31) P = 0.04)
Nersallah ( ASAIO J
2008)
RCT
A multicenter cross-over RCT
A = conventional HD
Randomisation
B = Blood Volume
tracking HD
A
B
A
B
B
A
B
A
Sequence 1
A
patients
36
0
2
Run In
33
32
3 protocol
violators
1 drop out
6
10
14
Sequence 2
18
time (weeks)
Experimental phase
Santoro A. et al., Kidney Int 2002
30% reduction in IDH during BV tracking HD
Santoro A. et al, Blood volume controlled hemodialysis in hypotension-prone patients: A
randomized, multicenter controlled trial, Kidney Int. 2002, 62, 1034-1045
Dialysis efficiency
Less IDH in Biofeedback HD (p< 0.001)
Standard
Biofeedback
Single Pool Kt/V
1.34 ± 0.08
1.26 ± 0.06
p< 0.005
Equilibrated Kt/V
1.03 ± 0.08
1.12 ± 0.05
p< 0.001
Urea Rebound %
14.2 ± 2.7
6.4 ± 2.3
p< 0.001
Urea removal (grams)
30.4 ± 4.1
35.4 ± 3.7
p< 0.005
Solute Removal Index
1.77 ± 0.15
2.01 ± 0.23
p< 0.005
Bland-Altman test (N=144 dialysis sessions)
Impact of biofeedback-induced cardiovascular stability on hemodialysis tolerance and efficiency
C. Ronco et al, Kidney Int., 2000, 58: 800-808
Intradialytic hypotension
The number of dialysis complicated by hypotensions over
the total number of assessed dialysis
Santoro A. submitted
European Best Practice Guidelines 2007
EBPG guideline on hemodynamic instability
Guideline 3.1.2a Individualized, automatic BV control should be
considered as a second-line option in patients with refractory IDH
(Evidence level II).
Rationale
With blood volume controlled treatments, ultrafiltration rate and/or dialysate
conductivity are adjusted according changes in relative blood volume.
[…] Nevertheless, several randomized cross-over studies have shown a
reduction in IDH and intra-dialysis symptomatology with the use of
automatic blood volume feedback [1,2,4-6]. Moreover, one study showed an
increase in dialysis efficacy with the use of this approach, due to a reduction
in intra-dialytic interventions [1].
[…] No adverse effects on sodium balance have yet been reported [2,7].
[…] Summarizing, various studies have shown a beneficial effect of
automatic blood volume controlled feedback in the prevention of IDH
episodes.
NDT (2007) 22 [Suppl. 2]: ii22-ii44
Measurement of blood volume during hemodialysis is a
useful tool to achieve safety adequate dry weight by
enhanced ultrafiltration
Method: N=12
Results:
• Single dry weight reduction the
mid-week dialysis = -0.5 Kg
• 58 % of patients were successful
(no symptoms)
• BV reduction recording
• 42 % of patients failed
(symptoms)
sensitivity
Receiver Operating Characteristic
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Fluid overload ?
∆BV/UF=2.8
∆BV/UF=2.6
∆BV/UF=2.4
∆BV/UF=2.2
0
0.1
0.2
0.3
0.4
1-specificity
Zellweger M et al, ASAIO J, 2004, 50, 242 - 245
∆BV/UF=3.0
0.5
0.6
• Prospective, randomized, parallel group study with two arms
(standard HD vs BVT)
• Study duration: 4 wk Run-in + 12 weeks
• Enrolled patients = 28 (14 per arm)
• Hypertensive pts (pre-HD and/or post-HD BP >150/90 mmHg) in
antihypertensive treatment or with cardiotoracic ratio >0,5
• No intervention were designed in each arm to reduce the dry
body weight but the judgment of the nephrologists according to
the overhydration status
Standard-HD
BVT
159.1±
±17.0
156.1±
±14.3
* p<0,01
On average a reduction of 0.7Kg in the dry
weight was observed in the first 3 weeks
166.1±
±19.8
143.6±
±23.3*
ECW/TBW
0,28
Standard HD
0,26
0,25
0,24 0,25
0,22
0
12
Weeks
BVT
ECW/TBW
0,28
0,26
p<0,01
0,24 0,25
0,23
0,22
0
12
Weeks
Conclusions
RBV monitoring has to be adjusted for UF rate and
weight in determing BV and hydration status
BV controlled HD (Hemo-biofeedback systems) proved
useful to improve the hemodynamic stability and the
overall tolerance to the HD treatment in DH-prone
population.
Frail, critical, co-morbid patients are the target patient
population for BV controlled HD.
EBPG consider automatic BV control in the strategies to
prevent hemodynamic instability.
Furthermore BV controlled HD has been shown to be
useful in the assessment of IBW in hypertensive pts.
with latent over-hydration.
Factors influencing KT/V
K
T
V
filter type, priming, Qb, clotting,….
by-passes, blood pump stops
dry weight, hydration status
Problems with conventional adequacy assessment
V- based dose measure may result in underdialysis of
women/ children/smaller patients
May fail to detect marked underdialysis if postBUN
drawn incorrectly
Expense of monthly postBUN blood drawn
Once-a-month measurement may not reflect montly
treatment (shortened, missed treatments)
• Large month-to-month variability
Current on-line adequacy
methods
Estimation of on-line dialyzer clearance
using sodium conductivity
Measuring or estimating the change in
spent dialysate urea during the treatment
Advantages of automated
monitoring of K or urea removal
Elimination of pre- and post-dialysis blood
urea nitrogen measurement
Ensuring that the patient receives the
prescribed dose of dialysis each time
More accurate delivery of a dialysis
prescription to new patients
Detection of access recirculation
Performing quality assurance of
reprocessed dialyzers
The aims of on-line monitoring
Keeping under continuous control physiological,
biochemical and haemodynamic parameters.
Preventing critical clinical situations
Modifying, in open-loop or with automatic feedback
(closed loop) the dialysis actuators
Limiting factors in intradialytic
on.line monitoring
Extra costs
Plentifully signals and poor knowledge
Polyedral interpretation
Larger validation studies in different
groups of patients may be needed so as
to evauated actual outcome effect